Potential symptom management targets following transarterial chemoembolization in patients with hepatocellular carcinoma: a network analysis and computer-simulation study
Background Potential symptom intervention targets in patients with hepatocellular carcinoma (HCC) after transarterial chemoembolization (TACE) remain poorly understood. Symptom network analysis combined with computer-simulated interventions may help identify priority targets for symptom management. Objective To examine symptom associations and explore potential symptom management targets among patients with HCC following TACE. Methods A cross-sectional study included 312 patients with HCC. Symptoms were assessed 24–48 h after TACE using the Chinese versions of the M. D. Anderson Symptom Inventory and the primary liver cancer-specific symptom module. A Gaussian graphical model was used to examine symptom associations, while an Ising model and NodeIdentifyR algorithm were used to explore simulated symptom perturbations. Results Fatigue (86.5%), disturbed sleep (83.7%), and lack of appetite (75.0%) were the most prevalent symptoms. Nausea showed the highest estimated strength centrality. Hypothetical alleviation of dry mouth produced the largest model-predicted decrease in the mean number of active symptoms, whereas hypothetical aggravation of fever produced the largest increase. Conclusions Nausea was the most connected symptom according to estimated strength centrality. Dry mouth and fever showed the largest model-predicted responses to alleviating and aggravating threshold perturbations, respectively. These findings are hypothesis-generating and require prospective validation; they do not establish clinical treatment effects.
Authors
- Lilan Yang
- Xudong Liu
- Zhuqing Zhang
- Santing Li
- Guifeng Li
- Xiangxiang Chen
Institutions
- Guangxi University of Chinese Medicine (CN)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-06
- DOI
- https://doi.org/10.3389/fmed.2026.1943199
- Primary Topic
- Cancer survivorship and care
- Type
- article
- Field-Weighted Citation Impact
- 0.00